Related papers: Tatemae: Detecting Alignment Faking via Tool Selec…

Highly capable AI systems could secretly pursue misaligned goals -- what we call "scheming". Because a scheming AI would deliberately try to hide its misaligned goals and actions, measuring and mitigating scheming requires different…

As Large Language Models (LLMs) expand in capability and application scope, their trustworthiness becomes critical. A vital risk is intrinsic deception, wherein models strategically mislead users to achieve their own objectives. Existing…

Mechanistic approaches to deception in large language models (LLMs) often rely on "lie detectors", that is, truth probes trained to identify internal representations of model outputs as false. The lie detector approach to LLM deception…

Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse applications, yet they pose significant security risks that threaten their safe deployment in critical domains. Current security alignment methodologies…

Aligning AI systems with organizational decision-making is typically framed as a single-target problem: make the model behave like the organization. We argue this framing obscures a deeper pluralistic challenge. We rely on a decision-policy…

Large Language Models increasingly power critical infrastructure from healthcare to finance, yet their vulnerability to adversarial manipulation threatens system integrity and user safety. Despite growing deployment, no comprehensive…

Code contains security and functional bugs. The process of identifying and localizing them is difficult and relies on human labor. In this work, we present a novel approach (FLAG) to assist human debuggers. FLAG is based on the lexical…

Large language models (LLMs) are increasingly used in human-AI interaction research and practice, yet existing capability and safety benchmarks reveal little about the value priorities these systems express or how those priorities…

Reinforcement Learning from Human Feedback (RLHF) assumes annotator preferences reflect stable internal states. We challenge this through three experiments spanning the preference pipeline. In a human choice blindness study, 91% of…

Tool-using LLM agents fail through trajectories rather than only final responses, as they may execute unsafe tool calls, follow injected instructions, comply with harmful requests, or over-refuse benign tasks despite producing a seemingly…

Large language models (LLMs) rely on safety alignment to avoid responding to malicious user inputs. Unfortunately, jailbreak can circumvent safety guardrails, resulting in LLMs generating harmful content and raising concerns about LLM…

Large language models (LLMs) have shown remarkable capabilities in various natural language processing tasks, yet they often struggle with maintaining factual accuracy, particularly in knowledge-intensive domains like healthcare. This study…

Federated Learning (FL) is a paradigm in Machine Learning (ML) that addresses data privacy, security, access rights and access to heterogeneous information issues by training a global model using distributed nodes. Despite its advantages,…

Anomaly detection in computational workflows is critical for ensuring system reliability and security. However, traditional rule-based methods struggle to detect novel anomalies. This paper leverages large language models (LLMs) for…

Large Language Models (LLMs) push the bound-aries in natural language processing and generative AI, driving progress across various aspects of modern society. Unfortunately, the pervasive issue of bias in LLMs responses (i.e., predictions)…

Aligning Large Language Models (LLMs) is crucial for enhancing their safety and utility. However, existing methods, primarily based on preference datasets, face challenges such as noisy labels, high annotation costs, and privacy concerns.…

Large language models (LLMs) are now ubiquitous in everyday tools, raising urgent safety concerns about their tendency to generate harmful content. The dominant safety approach -- reinforcement learning from human feedback (RLHF) --…

Logic locking refers to a set of techniques that can protect integrated circuits (ICs) from counterfeiting, piracy and malicious functionality changes by an untrusted foundry. It achieves these goals by introducing new inputs, called key…

Large language models (LLMs) are increasingly deployed in enterprise settings where they interact with multiple users and are trained or fine-tuned on sensitive internal data. While fine-tuning enhances performance by internalizing domain…

The rapid adoption of generative AI tools has heightened concerns regarding academic integrity, as students increasingly engage in dishonest practices by copying or paraphrasing AI-generated content. Existing plagiarism detection systems,…